Summary of Distribution Learnability and Robustness, by Shai Ben-david et al.
Distribution Learnability and Robustness
by Shai Ben-David, Alex Bie, Gautam Kamath, Tosca Lechner
First submitted to arxiv on: 25 Jun 2024
Categories
- Main: Machine Learning (stat.ML)
- Secondary: Data Structures and Algorithms (cs.DS); Information Theory (cs.IT); Machine Learning (cs.LG); Statistics Theory (math.ST)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper investigates the relationship between learnability and robust (agnostic) learnability in distribution learning settings. The authors surprisingly find that realizable learnability of a probability distribution class does not guarantee its agnostic learnability. They delve deeper, exploring types of data corruption that can disrupt distribution learnability and what makes it robust against certain corruptions. Notably, they show that realizable learnability implies robust learnability against additive corruption but not subtractive corruption. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper looks at how well we can learn patterns in a set of random events. The researchers found that just because we can learn something about the pattern doesn’t mean we can still learn it even if some data is changed or removed. They also studied what kind of changes to the data can make it harder to learn the pattern and what makes our learning method strong against certain types of changes. |
Keywords
* Artificial intelligence * Probability
